Bug 1034880

Summary:	[RFE] Stopping Gluster Volume while Quotas are Enabled Causes a Stale Mount
Product:	[Red Hat Storage] Red Hat Gluster Storage	Reporter:	Aron Gunn <agunn>
Component:	glusterfs	Assignee:	Raghavendra G <rgowdapp>
Status:	CLOSED DUPLICATE	QA Contact:	Sudhir D <sdharane>
Severity:	medium	Docs Contact:
Priority:	medium
Version:	2.1	CC:	agunn, saujain, vagarwal, vbellur
Target Milestone:	---	Keywords:	FutureFeature
Target Release:	---
Hardware:	x86_64
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Enhancement
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2013-11-28 04:08:00 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Aron Gunn 2013-11-26 16:17:44 UTC

Description of problem:
When stopping the Gluster volume before disabling quota on the volume will generate a 'df' error.

Version-Release number of selected component (if applicable):
glusterfs-server-3.4.0.35rhs-1.el6rhs.x86_64

How reproducible:
Consistently

Steps to Reproduce:
-Volume layout:
[root@rhsn1 ~]# gluster volume info
 
Volume Name: dist_test_vol01
Type: Distribute
Volume ID: d7aeb433-4b3e-4c29-ae09-bf956f11c10d
Status: Started
Number of Bricks: 2
Transport-type: tcp
Bricks:
Brick1: rhsn1:/rhs/brick01/vol01
Brick2: rhsn3:/rhs/brick01/vol01

-Enable quota on a volume:
[root@rhsn1 ~]# gluster volume quota dist_test_vol01 enable
volume quota : success

Volume Name: dist_test_vol01
Type: Distribute
Volume ID: d7aeb433-4b3e-4c29-ae09-bf956f11c10d
Status: Started
Number of Bricks: 2
Transport-type: tcp
Bricks:
Brick1: rhsn1:/rhs/brick01/vol01
Brick2: rhsn3:/rhs/brick01/vol01
Options Reconfigured:
features.quota: on

-A local FUSE mount is created:
[root@rhsn1 ~]# df -Ph
Filesystem            Size  Used Avail Use% Mounted on
...
localhost:dist_test_vol01 1015M  534M  482M  53% /tmp/dist_test_vol01

[root@rhsn1 ~]# mount
...
localhost:dist_test_vol01 on /tmp/dist_test_vol01 type fuse.glusterfs (rw,default_permissions,allow_other,max_read=131072)

-Stop the volume:
[root@rhsn1 ~]# gluster volume stop dist_test_vol01
Stopping volume will make its data inaccessible. Do you want to continue? (y/n) y
volume stop: dist_test_vol01: success

-Output from 'df':
[root@rhsn1 ~]# df -Ph
Filesystem            Size  Used Avail Use% Mounted on
...
df: `/tmp/dist_test_vol01': Transport endpoint is not connected

Actual results:
No warning when stopping the Gluster volume.
df: `/tmp/dist_test_vol01': Transport endpoint is not connected

Expected results:
Provide a warning that if the Gluster volume is stopped (while quotas are enabled), the quota features may not work properly, but still stop the Gluster volume.  Maybe something similar to a busy mount point warning.

Comment 1 Saurabh 2013-11-27 06:16:20 UTC

A similar BZ 1026227 already exists.
How about moving this bz to be a duplicate of the one already exists?

Comment 2 Vivek Agarwal 2013-11-27 07:36:53 UTC

Close the other one as duplicate, reason being this has a customer case attached to it.

Comment 3 Raghavendra G 2013-11-27 11:14:11 UTC

Aron,

May be we can unmount the auxiliary mount point while stopping the volume and remount once we start it. Does it sound reasonable?

regards,
Raghavendra.

Comment 4 Sudhir D 2013-11-28 04:08:00 UTC


*** This bug has been marked as a duplicate of bug 1026227 ***

Comment 5 Aron Gunn 2014-06-09 14:31:49 UTC

(In reply to Raghavendra G from comment #3)
> Aron,
> 
> May be we can unmount the auxiliary mount point while stopping the volume
> and remount once we start it. Does it sound reasonable?
> 
> regards,
> Raghavendra.

Raghavendra,

I know it's closed now and sorry for the extended delay in my response.  Yes, that sounds reasonable.

Thanks,
Aron